[Gllug] Alexandria and bar code scanners

[James Courtier-Dutton]

2009/8/10 John G Walker <johngeoffreywalker at yahoo.co.uk>:
>
>
> On Mon, 10 Aug 2009 17:43:16 +0100 JLMS <jjllmmss at googlemail.com> wrote:
>
>> Unless you have some interesting statistics, such denial is highly
>> counter intuitive...
>
> I don't have statistics, but I was objecting to a wildly broad
> statement that, itself, was backed up neither by statistics nor any
> qualitative evidence. Maybe you should ask James if he has statistics
> to back up his - extraordinarily wide - claim.
>
Industry stats:
a well-trained data entry operator will usually make a data entry
error once every 300 keystrokes.
Barcode Type 	Worst Case Accuracy 	Best Case Accuracy
Data Matrix       1 error in 10.5 million 	1 error in 612.9 million
PDF417 	1 error in 10.5 million 	1 error in 612.4 million
Code 128 	1 error in 2.8 million 	1 error in 37 million
Code 39 	1 error in 1.7 million 	1 error in 4.5 million
UPC 	1 error in 394 thousand 	1 error in 800 thousand

An ISBN book barcode is normally of type UPC or similar.
So taking the worse case,
Person: 1 error in 300
UPC barcode: 1 error in 394 thousand.
Me thinks the barcode wins.
-- 
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug